The development of drug resistance in leukemia cells constitutes the major reason for treatment failure in acute myeloid leukemia (AML). Although a number of prognostic factors such as cellular morphology and cytogenetic abnormalities have been identified, the mechanisms by which cellular resistance to chemotherapy occurs are poorly understood. Because even unsuccessful regimens produce substantial cytoreduction, it is likely that characteristics of subpopulations of leukemic cells will be of importance in the development of resistance. The University of Maryland Cancer Center (UMCC) will perform multiple laboratory studies on leukemic blasts to simultaneously evaluate a number of proposed mechanisms in vitro. These results will be correlated with outcome after treatment and clinical parameters in a consecutive cohort of newly diagnosed patients treated on clinical trials. Serial evaluations will be done at the time of treatment failure or relapse. We propose to focus on: cellular and pharmacokinetic mechanisms of resistance of daunorubicin with assessment of the multidrug resistance phenotypes associated with p-glycoprotein and topoisomerase II abnormalities in leukemia cells; evaluation of methods for determination of drug sensitivity; in vitro and subsequent in vivo evaluation of non- cytotoxic modulaters of p-glycoprotein function. Flow cytometry, molecular hybridization and immunologic techniques will be utilized to identify these various parameters in subpopulations of leukemic cells. The UMCC has a large population of leukemia patients, extensive preliminary data on the proposed projects, a history of productive clinical-laboratory interactions and represents an excellent environment in which to study these important questions.